lkml.org 
[lkml]   [1997]   [Feb]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: Unable to handle kernel paging request
    I just ran the program Bernd posted.  It segfaulted, and I got a similar
    looking error.

    Unable to handle kernel paging request at virtual address 40008000
    current->tss.cr3 = 013a3000, %cr3 = 013a3000
    *pde = 02347067
    *pte = 00000000
    Oops: 0000
    CPU: 0
    EIP: 0010:[<00157b45>]
    EFLAGS: 00010216
    eax: 0217fd04 ebx: 00000400 ecx: 00000100 edx: 02947800
    esi: 40008000 edi: 02947800 ebp: 00000400 esp: 02994c08
    ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018
    Process tickle (pid: 30809, process nr: 17, stackpage=02994000)
    Stack: 00001000 40008000 40008000 00040000 00000000 00000000 001d9148
    00000000
    00000000 00000014 00005000 00000000 0217fd04 ffffffe4 021b6d90
    00ee3b6c
    02994c94 01a8ad90 00008180 02994c94 01a8ad90 00008180 00123d52
    00ee3b6c
    Call Trace: [<00123d52>] [<0015ab26>] [<0012ef24>] [<0012f05b>] [<0012ef24>]
    [<0012fa34>] [<00118dc3>]
    [<00118e47>] [<00118cc0>] [<0010a42f>] [<0010a672>]
    Code: 64 f3 a5 83 e3 03 89 d9 64 f3 a4 55 8b 54 24 34 8b 52 24 03

    Using `/kernels/System.map-2.0.27-fdt-elf' to map addresses to symbols.

    >>EIP: 157b45 <ext2_file_write+249/45c>
    Trace: 123d52 <__brelse+22/44>
    Trace: 15ab26 <ext2_create+152/168>
    Trace: 12ef24 <dump_write+1c/2c>
    Trace: 12f05b <writenote+a7/c8>
    Trace: 12ef24 <dump_write+1c/2c>
    Trace: 12fa34 <elf_core_dump+9b8/a50>
    Trace: 118dc3 <do_no_page+103/328>
    Trace: 118e47 <do_no_page+187/328>
    Trace: 118e47 <do_no_page+187/328>
    Trace: 10a42f <do_signal+1ef/280>
    Trace: 10a672 <signal_return+12/40>

    Code: 157b45 <ext2_file_write+249/45c> repz movsl %ds:(%esi),%es:(%edi)
    Code: 157b48 <ext2_file_write+24c/45c> andl $0x3,%ebx
    Code: 157b4b <ext2_file_write+24f/45c> movl %ebx,%ecx
    Code: 157b4d <ext2_file_write+251/45c> repz movsb %ds:(%esi),%es:(%edi)
    Code: 157b50 <ext2_file_write+254/45c> pushl %ebp
    Code: 157b51 <ext2_file_write+255/45c> movl 0x34(%esp,1),%edx
    Code: 157b55 <ext2_file_write+259/45c> movl 0x24(%edx),%edx
    Code: 157b58 <ext2_file_write+25c/45c> addl (%eax),%eax
    Code: 157b5a <ext2_file_write+25e/45c> nop
    Code: 157b5b <ext2_file_write+25f/45c> nop
    Code: 157b5c <ext2_file_write+260/45c> nop

    I then ran it a second time, to see if I'd get identical results, and
    instead I got a process eating 90% CPU that I cannot kill. Running it a
    third time left the 3rd invocation in uninteruptable sleep.

    fubar 30823 89.5 0.3 820 184 ? R 11:31 11:46 ./tickle
    fubar 31312 0.1 1.1 1204 564 p5 S 11:42 0:00 -bash
    fubar 31359 0.0 0.3 820 184 p5 D 11:44 0:00 ./tickle

    Guess it's time to upgrade kernels and reboot.

    On Mon, 24 Feb 1997, Bernd Schmidt wrote:

    > > > eax: 00986bb0 ebx: 00000400 ecx: 00000100 edx: 016bd400
    > > > esi: 40007000 edi: 016bd400 ebp: 00000400 esp: 01384c08
    > >
    > > Flipped 14th bit in esi, most likely. In other words, memory error.
    >
    > I'd like to agree, but I've seen another bug report like this, and I noticed
    > that in both cases, the function elf_core_dump appeared in the stack trace.
    > I even still have the original mail. The author (Johnny Stenback) claims the
    > problem is reproducible with the following test program:
    >
    > #include <stdio.h>
    > #include <unistd.h>
    > #include <sys/mman.h>
    > #include <sys/types.h>
    > #include <sys/stat.h>
    > #include <fcntl.h>
    > #include <string.h>
    > #include <errno.h>
    >
    > void main ()
    > {
    > int fd;
    > void *p;
    > char buf[1024];
    > fd = open("test", O_RDONLY);
    > p = mmap(NULL, 200, PROT_WRITE, MAP_PRIVATE, fd, 0);
    > strncpy (buf, p, 100);
    > printf ("%s\n", buf);
    > munmap (p, 200);
    > exit (0);
    > }
    >
    > The "Oops" from the other post is mostly identical-looking.
    > (I haven't tried this, so if it doesn't work, I apologize. I'll try today)
    > Of course, you'll have to enable core dumps.
    >
    > Bernd
    >



    ------------------------------------------------------------------
    Jon Lewis <jlewis@fdt.net> | Unsolicited commercial e-mail will
    Network Administrator | be proof-read for $199/hr.
    ________Finger jlewis@inorganic5.fdt.net for PGP public key_______


    \
     
     \ /
      Last update: 2005-03-22 13:39    [W:0.025 / U:1.472 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site